JST > Basic Research Programs > ERATO
Exploratory Research for Advanced Technology 
TOP Contact Us Site Map 日本語
HOME Outline of ERATO Research Projects Mid-term/Ex-post Evaluation Application
HOME > Research Projects > Ongoing Projects > SHIMOJO lmplicit Brain Function > Implicit auditory
Implicit auditory
Research Strategy
 1) A motor theory approach to emotion transmission: In the research of vocal communication, only the transmission of linguistic content has tended to be research objectives, but the characteristics of the so-called non-linguistic content like voice rhythm and quality plays an essential role in transmitting emotion. How can one transmit emotions using the voice? For this question, I propose “a motor theory on emotion transmission” in which “a listener decodes a speaker’s emotion by adjusting the speaker’s voice to his own motor model of the utterance organ”. This can be said to be an emotion version or non-linguistic content version of the “motor theory of voice perception” regarding linguistic content which was proposed in 1960’s, and is recently coming back brightly after the discovery of the mirror neuron. In this group, we inspect the motor theory on emotion transmission experimentally. Specifically we are going to identify the neural correlates of the generation and perception of emotional voice using brain imaging methods, especially the side of the frontal lobe, opposite of Brocca’s area. We will measure and observe the motion of the utterance organ with electromyography preceding to awareness, and verify relevant neurons and the internal model decoding and operating emotion occurred from external stimulus. We will also construct an emotion transmission theory based on sensori-motor reciprocal action by inspecting the commonalities of emotional cognitive mechanisms between expression and gesture in visual perception.

2) Predictability of time-ordered sound events and emotion: Familiarity and novelty, or predictability and deviation, these are important factors of a pleasant music listening. If one can not find any predictability in a sequence of sound events, one feels dubious, yet if one can predict it completely, one feels boredom. It is thought that various forms and techniques in various music styles are developed to realize the most suitable compromise of both. As the mechanism ruling on predictability of time-ordered sound events, we propose the following hypotheses. Familiarity and novelty can not exist at the same time, but both of them essentially cause pleasure. The reason why familiarity makes pleasure might be reward to hit the prediction made by the internal model that might lead to reinforce it and become more accurate. On the other hand, due to detecting information of the revising internal model, novelty--namely a prediction error of the internal model--might evoke pleasure, which becomes more precise in the results from the internal model. Two neuronal networks of cortex level and under cortex level are possibly related to familiarity and novelty. A network under cortex might be a simple model (adaptive filter) based on statistical structure of closest stimuli, react extremely fast, make particular orienting after detecting novelty and give way to explicit process. A cortical network is thought to have a complex structure and react precisely but slowly. Reaction and emotion after hearing sounds would be prescribed by a timing and balance of these networks’ activity. One example is reaction for music. By this study, we inspect the hypothesis using brain imaging methods among other techniques.
(This group is located mainly in NTT Atsugi Research and Development Center.)

Staff Members
Group Leader Makio Kashino
Researcher Rodrigo Segnini
Research Assistant Minae Okada
Address of
Research Laboratory
NTT Atsugi Research and Development Center 3-1,
Morinosatowakamiya, Atsugi-shi,Kanagawa 243-0198, Japan
Phone : 046-290-5615
FAX : 046-248-2911

JST 過去のお知らせはこちら ERATO
Exploratory Research for Advanced Technology 
創造科学技術推進事業 ERATO